See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/319357436
On the surface structure of 4-mercaptopyridine on Au(111): A new dense phase
Article in Langmuir · August 2017
DOI: 10.1021/acs.langmuir.7b01627
CITATIONS
READS
3
212
6 authors, including:
Santiago Esteban Herrera
Federico Tasca
INIFTA-CONICET-UNLP
University of Santiago, Chile
19 PUBLICATIONS 114 CITATIONS
40 PUBLICATIONS 907 CITATIONS
SEE PROFILE
SEE PROFILE
Federico J. Williams
Ernesto J Calvo
Universidad de Buenos Aires
Universidad de Buenos Aires
115 PUBLICATIONS 2,011 CITATIONS
193 PUBLICATIONS 4,927 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
New nano/microfabrication methods View project
Fundamental studies of oxygen electroreduction reaction in non-aqueous electrolytes for the development of Lithium-air battery technology View project
All content following this page was uploaded by Santiago Esteban Herrera on 15 September 2017.
The user has requested enhancement of the downloaded file.
Article
pubs.acs.org/Langmuir
Surface Structure of 4‑Mercaptopyridine on Au(111): A New Dense
Phase
Santiago Herrera, Federico Tasca,† Federico J. Williams, and Ernesto J. Calvo*
Departamento de Química Inorgánica, Analítica y Química Física, INQUIMAE-CONICET, Facultad Ciencias Exactas y Naturales,
Universidad de Buenos Aires, Ciudad Universitaria, Pabellón 2, Buenos Aires C1428EHA, Argentina
Pilar Carro
Á rea de Química Física, Departamento de Química, Facultad de Ciencias, Universidad de La Laguna, Instituto de Materiales y
Nanotecnología, Avenida Francisco Sánchez, s/n 38200-La Laguna, Tenerife, Spain
Roberto C. Salvarezza
INIFTA Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas (INIFTA), Facultad de Ciencias Exactas, Universidad
Nacional de La Plata - CONICET- Sucursal 4 Casilla de Correo 16, 1900) La Plata, Argentina
ABSTRACT: 4-Mercaptopyridine (4MPy) self-assembled on
Au(111) has been studied by in situ electrochemical scanning
tunneling microscopy (EC-STM) in HClO4, cyclic voltammetry, Xray photoelectron spectroscopy (XPS) and density functional
theory (DFT). Samples prepared by varying the immersion time at
constant concentration named short time (30 s) and long time (3
min) adsorption have been studied. Cyclic voltammetry and XPS
showed that the chemistry of the adsorbed molecules does not
depend on the adsorption time resulting in a well established
chemisorbed thiol self-assembled monolayer on Au(111). EC-STM
study of the short time adsorption sample revealed a new selfassembled structure after a cathodic desorption/readsorption
sweep, which remains stable only if the potential is kept negative
to the Au(111) zero charge potential (EPZC). DFT calculations have
shown a correlation between the observed structure and a dense weakly adsorbed phase with a surface coverage of θ = 0.4 and a
(5 × √3) lattice configuration. At potentials positive to the EPZC, the weakly adsorbed state becomes unstable, and a different
structure is formed due to the chemisorption driven by the electrostatic interaction. Long time adsorption experiments, on the
other hand, have shown the typical (5 × √3) structure with θ = 0.2 surface coverage (chemisorbed phase) and are stable over
the whole potential range. The difference observed in long time and short time immersion can be explained by the optimization
of molecular interactions during the self-assembly process.
■
0.20/0.25 with the formation of gold islands.4 In this case, it is
not clear if RS−Auad−RS or simply RS-Au adsorbates are
formed. In particular, scanning tunneling microscopy (STM)
has provided a detailed description of the different phases
formed along the adsorption time that involve the initial
formation of lying down phases (aliphatic)5 or highly tilted
phases (aromatic)6 followed by molecular organization in
denser standing up configurations.5 By contrast, little is known
about the physisorbed state of these molecules on the Au(111)
substrate for which we lack structural data.
INTRODUCTION
Self-assembled monolayers (SAMs) of thiols on Au(111) have
been extensively studied due to their multiple applications in
nanotechnology and also as model systems to understand
molecular adsorption on metal substrates.1 At present it is
accepted that thiols (RSH) first physisorb on the Au surface
without S−H bond scission and later chemisorb forming a
strong RS−Au bond, known as thiolates, and hydrogen with
simultaneous lifting of the (22 × √3) surface reconstruction.2
Today we have a good understanding of the chemisorption
process for aliphatic thiols with a surface coverage θ= 0.33
accompanied by the formation of gold vacancy island and RS−
Auad−RS (Auad = Au adatom) moieties that are the basic units
of the alkanethiol SAMs.3 On the other hand, for small
aromatic and heterocyclic thiols, this process results in θ =
© XXXX American Chemical Society
Received: May 16, 2017
Revised: August 26, 2017
Published: August 29, 2017
A
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Electrochemical desorption measurements were carried out with an
Autolab V 30 system (Eco Chemie, Utrecht, The Netherlands)
controlled by NOVA 2 Software. A custom-made three-electrode
Teflon cell exposing an area of 0.196 cm2 was used to hold the gold
single crystal. The reference electrode used was AgCl/Ag. For the
current density calculation, the electroactive area of the electrode was
obtained by integrating the gold reduction peak in the cyclic
voltammetry in 0.1 M HClO4.
X-ray photoelectron spectroscopy (XPS) measurements were
conducted in an ultrahigh vacuum (UHV) chamber with a base
pressure below 5 × 10−10 mbar, using a 150 mm hemispherical SPECS
electron energy analyzer and a monochromatic Al Kα X-ray source.
Binding energies reported in this work are referenced to the Fermi
edge of Au(111) at EB = 0 eV. Atomic ratios were calculated from the
integrated intensities of core levels after instrumental and photoionization cross section corrections.
Density functional calculations were performed with the periodic
plane-wave basis set code VASP 5.2.12.14,15 We have followed the
scheme of nonlocal functional proposed by Dion et al.,16 vdW-DF, and
the optimized Becke88 exchange functional optB88-vdW17 to take into
account van der Waals (vdW) interactions. The electronic wave
functions were expanded in a plane-wave basis set with a 420 eV cutoff
energy. The projector augmented plane wave (PAW) method has been
used to represent the atomic cores with PBE potential.18 Optimal grid
of Monkhorst−Pack19 k-points of 2 × 9 × 1 has been used for
numerical integration in the reciprocal space of the (5 × √3) surface
structure and we have only considered the Γ point for the (22 × √3)
surface structure. Both neutral tautomers 4-mercaptopyridine
(4MPyHS) and 1H-pyridine-4-thione (4MPyNH) and 4MPy(S*)
radical species were optimized in an asymmetric box of 14 Å × 16
Å × 18 Å. The projector augmented wave (PAW) method,18,20 as
implemented by Kresse and Joubert,21 was employed to describe the
effect of the inner cores of the atoms on the valence electrons. The
energy minimization (electronic density relaxation) for a given nuclear
configuration was carried out using a Davidson block iteration scheme.
The dipole correction was applied to minimize polarization effects
caused by asymmetry of the slabs.
The surface was modeled by a periodic slab composed of five metal
layers for the (5 × √3) surface structure and only three layers for the
(22 × √3) due to the large size of the unit cell. In both unit cells, the
vacuum layer was of ∼14 Å. Adsorption occurs only on one side of the
slab. During the geometry optimization the two bottom layers (5 ×
√3) and only one (22 × √3) were kept fixed at their optimized bulk
truncated geometry for the Au(111) surface. The three (5 × √3) or
two (22 × √3) outermost atomic metal layers, as well as the atomic
coordinates of the adsorbed species, were allowed to relax without
further constraints. The atomic positions were relaxed until the force
on the unconstrained atoms was less than 0.03 eV/ Å and the
tolerance used to define the self-consistency was 10−5 eV for the
single point energy. The lattice parameter calculated for bulk Au was
4.16 Å, which compares reasonably well with the experimental value
(4.078 Å).
We define the average binding energy (Eb) per 4MPy species
adsorbed as follows
In situ STM under electrochemical control is an excellent
technique to study the effect of the applied potential of
adsorbed molecules on metals. This technique allows one to
detect potential induced phase transitions in adsorbed thiols as
the surface charge of the substrate is changed. One of the most
studied aromatic thiols on Au(111) by in situ STM has been 4mercaptopyridine (4MPy) in acid media. This interesting thiol,
which exposes an N heteroatom to the outer SAM interface,
allows one to perform different chemical reactions on the SAM
to built complex tridimensional architectures. It is well known
that 4MPy decomposes in neutral and alkaline solutions,
yielding atomic S on the Au surface, a process that takes
minutes to hours to complete but is stable in acid solutions.7
Therefore, STM data in acid media have shown that 4MPy
organizes in striped rectangular surface structures, namely, a (5
× √3) lattice,8−10 with a related (10 × √3) superstructure,11
and a (7 × √3) lattice.12 DFT calculations without van der
Waals interactions have shown that the (7 × √3) is more
stable than the (5 × √3).13 Interestingly, a more dense (1 ×
√3) periodicity has been also reported.11,12
In this work we have performed short time adsorption
experiments (t < 1 min) to form a 4MPy adlayer on Au(111)
from aqueous solutions. This adlayer has been placed in the
STM-electrochemical cell containing 0.1 M HClO4 to follow
potential induced transformations. EC-STM results reveal the
presence of a dense phase when the potential applied (E) to the
molecule-Au interface is more negative than the zero charge
potential (EPZC). We propose that this dense phase is weakly
adsorbed on the Au surface, stabilized by intermolecular π−π
interactions and electrostatics forces with the negatively
charged substrate, and evolves into a new closely related lattice
at E ≥ EPZC. This intermediate lattice can be transformed into
the well-known (5 × √3). When the adsorption time is
increased to t > 2 min, only the (5 × √3) lattice of
chemisorbed 4MPy molecules is observed, irrespective of the
applied potential, i.e., the weakly adsorbed state is not observed.
Our results demonstrate the existence of dense and organized
weakly adsorbed phases of thiols on Au(111), allowing for a
complete description and connections between the different
surface structures in terms of thermodynamic stability.
■
EXPERIMENTAL SECTION
4-Mercaptopyridine (95%, Aldrich) and HClO4 solution (ACS
reagent, 60%, Aldrich) were used without further purification. A
scanning probe microscope AFM-STM 5500 (AgilentTechnologies)
with four-electrode bipotentiostat for the independent control of
substrate and tip potentials with respect to the reference electrode in
the electrolyte was used for in situ electrochemical scanning tunneling
microscopy (EC-STM) measurements. The gold single-crystal
electrode used was a solid cylinder (MaTeck, Germany, 1 cm
diameter, polished down to 0.03 μm) with one of its ends oriented to
better than 1° along the (111) plane. Prior to each experiment, the Au
crystal was annealed for 5 min in a hydrogen flame and allowed to cool
in air. Tungsten tips were made by electrochemical etching in 2 M
KOH and then cleaned with concentrated hydrofluoric acid, water, and
acetone. To minimize Faradaic currents, the tips were isolated with
nail paint and dried overnight. The custom-made three-electrode
PTFE cell was used with two Pt wires used as counter and
pseudoreference electrodes, respectively (all potentials herein are
referred to Ag/AgCl reference electrode). An HClO4 0.1 M aqueous
solution was used as supporting electrolyte.
The thiol adsorption was performed by immersion of a clean
Au(111) substrate in a freshly prepared 0.27 mM 4MPy aqueous
solution (3 mg in 100 mL of Milli-Q water) between 30 s and 3 min
depending on the experiment.
Eb =
1
[Etotal − EAu(111) − E@]
N@
(1)
where N@ is the number of adsorbed species in the surface unit cell,
Etotal stands for the total energy of the system, EAu(111) is the energy of
the clean surface, and E@ is the energy of the adsorbate either neutral
or radical. Negative numbers indicate an exothermic adsorption
process with respect to the clean surface and the adsorbed phase
originated during the adsorption process.
The Gibbs free energy of adsorption of the surface structure (γ) was
approximated through the total energy from DFT calculations by using
eq 2:
Y=
B
N@E b
A
(2)
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. EC-STM images of 30 s adsorption of aqueous 4MPy on Au(111). Sample bias and set point current were fixed to +400 mV and 600 pA,
respectively. HClO4 0.1 M was used as electrolyte. The substrate potential was sequentially changed to (A) +0.330 V; (B) +0.03 V; (C) +0.230 V;
(D) +0.330 V. (E) STM image after scanning for 20 min starting from item D (+0.330 V); (F) + 0.380 V (Phase B); (G) STM image after scanning
for 5 min at +0.480 V; (H) STM image after scanning for 10 min starting from item G (+0.480 V); (I) +0.580 V (Phase C).
where A is the unit cell area. Considering that we are concerned with
free energy differences, it is reasonable to assume that the
contributions coming from the configurational entropy, the vibrations,
and the work term, pV, can be neglected.22,23
no islands could be observed, and some ordered domains began
to emerge from the completely disordered molecular background structure (Figure 1C). The ordering process is even
more evident returning to E = 0.33 V, where the molecules now
form dense stripes arrays (Figure 1D) whose density increases
either with time (Figure 1E) or by increasing the applied
potential up to E = 0.38 V (Figure 1F). Note, however, that an
important fraction of the substrate is not covered by the
ordered arrays of molecules (dark regions in Figure 1F). The
short-range order surface structure obtained at this point will be
named herein as phase B.
Increasing E to 0.48 V, a potential value where the (22 ×
√3) to (1 × 1) surface phase transition starts (0.49 V),24
dramatic changes in the molecular structure take place. First,
the density of ordered stripes decreases (Figure 1G,H), and
finally a new striped structure that we will call phase C with
long-range order is clearly visible at E = 0.58 V, in coexistence
with disordered islands (Figure 1I).
A more detailed analysis of the B phase, which is observed in
the 0.3−0.4 V potential region where the (22 × √3) is stable,
is shown in Figure 2A,B.
First, the molecular domains of this phase are relatively small
(<10 nm) with a short-range order as previously mentioned.
The self-assembled molecular stripes intersect themselves
forming 60° or 120° angles consistent with the underlying
substrate directions (Figure 2A). A closer examination of the
■
RESULTS AND DISCUSSION
Figure 1 shows typical in situ STM images in 0.1 M HClO4of 4MPy modified Au(111) surface by adsorption at short time.
Figure 1A was taken at open circuit potential (ocp), while
panels B to I depict the surface morphology at different
consecutive applied potentials. The ocp recorded at the
beginning of the experiment was Eocp= 0.33 V, a value close
the potential of zero charge of Au(111), EPZC= 0.36 V.24 In
Figure 1A, the image reveals Au terraces covered by disordered
arrays of molecules and randomly distributed islands over the
surface. The cross section analysis shows that the apparent
height of these features is 0.24 nm, which corresponds to Au
islands that could result from the lifting of the (22 × √3)
surface reconstruction induced by the 4MPy adsorption from
the 4MPy containing aqueous solution.
When the potential was shifted to E = 0.03 V, the Au islands
disappeared, suggesting that the 22 × √3 surface reconstruction is now the stable surface structure in this potential region
(Figure 1B). At this point, a stationary current density has been
recorded (−0.5 μA·cm2), indicating the onset of the hydrogen
evolution. If the surface potential is then moved to E= 0.23 V,
C
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 2. EC-STM images showing a 4MPy/Au(111) self-assembled
monolayer (30 s adsorption). (A) Phase B, Esub = +330 V, 50 nm × 50
nm scan; (B) Phase B, Esub = +330 V, 13 nm × 13 nm scan (3.75 nm
× 3.75 nm STM high resolution inset); (C) Phase C, Esub = +580 V,
130 nm × 130 nm scan; (D) Phase C, Esub = +580 V, 10 nm × 10 nm
scan. Ebias = +400 mV, Itunn = 600 pA. HClO4 0.1 M was used as
electrolyte.
Figure 3. EC-STM images showing a 4MPy/Au(111) phase A selfassembled monolayer (2.5 min adsorption). Esub = +0.390 V, Etip =
+0.03 V, Itunn = 600 pA. (A) 60 nm × 60 nm scan; (B) 15 nm × 15 nm
scan (3.1 nm × 3.1 nm STM high resolution inset). HClO4 0.1 M was
used as electrolyte.
phase B structure shows that the stripes are separated by 0.53 ±
0.03 nm (Figure 2B) while the distance between molecules is
0.32 ± 0.01 nm (inset in Figure 2B), leading to a surface
coverage of θ = 0.40. These structures resembles those
observed by in situ STM imaging of bipyridine molecules on
the Au(111) surface by Tao et al.25 The large density of
irregular holes (dark regions in Figure 2) exhibits 0.18 nm in
depth, indicating that they are uncovered substrate regions and
not vacancy gold pits since 0.24 nm would be expected in that
case.
On the other hand, the C phase is observed at 0.58 V (Figure
2C−D), a potential value where the stable structure of the
single crystal should be the Au(111)-(1 × 1). The STM image
reveals long-range molecular domains with sizes of about 10
nm. Typical distances between stripes and molecules were
found to be 1.32 ± 0.02 nm and 0.33 ± 0.02 nm, respectively.
Interestingly, the C phase has the same intermolecular distance
as the B phase but much larger interstripe separation.
If a new experiment is carried out, but in this case the
adsorption time is increased up to 2.5−3 min, the phase A is
obtained after a few scans at open circuit potential (Eocp=
+0.390 V). Note that in this case as in the previous one, the
concentration of the 4MPy aqueous solution was carefully fixed
to be 0.270 mM. Figure 3 shows that phase A exhibits longrange molecular domains with sizes of about 10−20 nm. Each
domain consists of stripes that also intersect themselves
following the underlying substrate directions as in the phase B.
Some disordered regions between the ordered domains are
also observed by STM. A few holes 0.24 nm in depth (dark
regions in the images) corresponding to vacancy gold islands
are also observed in the figure. The analysis of the characteristic
distances of the surface structure shows 1.44 ± 0.01 nm stripe
separation (Figure 3B) and 0.53 ± 0.02 nm intermolecular
distance inside the stripes. These figures correspond to the
well-known (5 × √3) lattice with θ = 0.20 already reported by
several authors in acid media.26,27 Interestingly, this phase is
now stable irrespective of the applied potential. In fact, this
phase was imaged in the whole potential range before reaching
a potential of E = +0.03 V where complete disorder of the
4MPy molecules was observed resulting in an apparent clean
Au surface (hydrogen evolution region). Figure 4A to 4D show
consecutive STM imaging of phase A starting from ocp, going
down to +0.03 V and then going up to +0.580 V (intermediate
images between +0.130 V and +0.580 V show the same feature
seen in Figure 4D).
In order to estimate the amount of 4MPy resulting from the
self-assembly process, reductive desorption measurements were
performed in alkaline solution after 4MPy adsorption for either
30 s or 3 min (Figure 5). In both cases, a charge density of q =
35 ± 5 μC cm−2 was obtained, a figure that for one electron
transfer results in θ = 0.15, closer to that expected for the
diluted (5 × √3) lattice. The presence of a hump at −0.78 V
can be assigned to 4MPy molecules adsorbed at step edges.
Figure 6 shows the C 1s, S 2p, N 1s, and O 1s XPS spectra
corresponding to the 30 s and 2.5 min respectively 4MPy
modified Au(111) surface. The C 1s spectra show a broad
signal centered at 284.9 eV corresponding to the different
carbon environments in the molecule. The S 2p spectra show
the expected doublet corresponding to the S 2p3/2 (162.1 eV)
and S 2p1/2 (163.3 eV) with a 2:1 intensity ratio. The binding
energy position of the sulfur peak and the absence of other
contribution in this spectral region indicate that all the
D
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 4. Thiol desorption and stability of phase A over a wide
potential window. (A) Initial scan at E = 0.390 V (ocp); (B) scan at E
= 0.130 V (after A); (C) thiol desorption at E = 0.03 V; (D) phase A
thiol readsorption at E = 0.130 V, same feature observed from E =
0.130 V to E = 0.580 V. Ebias = +400 mV, Itunn = 600 pA. Scan size 40 ×
40 nm. HClO4 0.1 M was used as electrolyte.
Figure 6. C 1s, S 2p, N 1s, and O 1s XPS spectra after exposing the
Au(111) surface to an aqueous solution of 4-mercaptopyridine for 30 s
and 2.5 min.
pyridine molecules.29,30 The different intensities of each
contribution differ in both cases due to the different degrees
of protonation in both cases (the pH of the rinsing solution was
not controlled). The XPS calculated C:N:S ratios are 5.2:1:0.96
and 5.1:1:1 for 30 s and 2.5 min, respectively. The excellent
agreement with the 5:1:1 stoichiometric ratio implies that the
molecules retain their molecular identity upon adsorption.
Finally the O 1s spectra show a very small contribution at
approximately 532.5 eV, presumably due to water molecules
attached to the pyridine molecules via H bonds.
Therefore, the XPS and electrochemical data indicates that
that no significant differences exist in the chemistry and the
surface coverage of the molecular adlayer in the 30 s to 3 min
time range used for the self-assembly process. Also, the absence
of a peak at −0.9 V and the 161 eV component in the S 2p
spectra, as well as the STM images where neither the typical
(√3 × √3)R30° and rectangular sulfur structures can be
observed, are clear evidence that no significant amounts of
sulfur contamination is present on the Au(111) surface.31
Now we will discuss the connection between diluted A and
dense B phases. While the A phase has been unambiguously
assigned to the (5 × √3), the surface structure of phase B
remains controversial. A dense (1 × √3) phase has been
reported for 4MPy at lower E values in H2SO4 media, but the
authors claim that the dense phase is never observed in HClO4
Figure 5. Reductive desorption of a 4MPy/Au(111) self-assembled
monolayer (at 30 and 2.5 s adsorption) in KOH 0.1 M. Scan rate: 20
mV·s−1.
molecules are chemisorbed via a Au−S bond28 after 30 s and
2.5 min of exposure to the aqueous 4MPy solution. The S/Au
ratio signal is similar in both samples with an estimate surface
coverage 0.15, in good agreement with the electrochemical
measurements. Here we should note that these are ex situ
measurements of the previously modified surface under no
electrochemical potential control. This implies that the dense
phase from the STM observations requires an electrochemical
bias in order to form.
The N 1s region shows three contributions. The lowest
binding energy signal centered at 398.5 eV is due to the
deprotonated N atoms in the pyridine molecule, whereas the
400.2 and 401.8 eV contributions are due to N atoms in
different degrees of H acceptance and protonation in the
E
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
media at any potential.11 In that case, a complex adlayer with
bisulfate anions coadsorbed on the protonated pyridinium N
atoms in order to decrease repulsive interactions has been
proposed. By contrast, the dense B phase presented in this
work is observed in HClO4 by playing with the molecule
adsorption time. Also, a dense phase with nearest neighbor
distances as small as 0.21 nm has been reported for 4MPy in
acid media, suggesting that the molecules are adsorbed as
disulfide species.32 However, the structures observed in these
previous works are inconsistent with the 0.32 nm distance
measured in our images which coincides with those expected
for molecular stabilization by π−π interactions between
aromatic heterocycles.
In order to understand the surface structure and chemistry of
the dense phase we have made DFT calculations for the wellknown (5 × √3) lattice with θ = 0.2 for the 4MPy* (Figure
7a) and compared them with dense (5 × √3) lattice phases of
4MPy* (Figure 7b) and of the thione tautomer (4MPyNH)
(Figure 7c) (θ = 0.4), both consistent with the STM
intermolecular distances. Also we have studied the adsorption
of the 4MPySH species on both the Au(1 × 1), (5 × √3), θ =
0.40 (Figure 7d) and (22 × √3), θ = 0.41 (Figure 7e,f) Au
surfaces. The optimized structures are shown in Figure 7a−f
and their corresponding geometrical and energetic parameters
are included in Table 1. First, the 4MPyNH species adsorbs
Table 1. Structural and Energetic Parameters for the 4MPy*,
4MPyHNH, and 4MPySH Models
(5 × √3)
surface lattice
(22 × √3)
adsorbate
4MPy*
4MPy*
4MPyNH
4MPySH
4MPySH
θ
Eb/eV
γ/meV·Å‑2
Z(S−Ausurf)/Å
α(N−Snormal_surface)/°
Bader
S
charge/e
Au
0.2
−2.10
−56.20
1.93
39.5
0.4
−2.04
−108.92
1.93
40.6
0.4
−0.85
−45.60
2.34
43.1
0.4
−0.90
−48.2
2.82
37.0
0.41
−0.81
−44.10
2.80/3.10
36.0
−0.09
+0.01
−0.11
+0.05
+1.2
−0.01
+1.33
−0.04
+1.32
0.00
close to the substrate but exhibits a small adsorption energy
(Eb) and less stable (less negative) surface free energy (γ) than
the diluted and dense 4MPy* species. On the other hand, the
4MPySH on both Au substrates exhibits similar low Eb and
lower γ values than the 4MPyNH species although it exhibits
much larger S−Au distances (≈3 Å) suggesting a physisorbed
state. Interestingly, while the diluted (5 × √3) chemisorbed
lattice exhibits much higher Eb than those of the 4MPyHNH and
4MPySH physisorbed lattices the difference in γ values is much
smaller (Δγ = 8/10 meV Å−2). The stability of these species at
(E < EPZC) arises from two contributions: (1) the π−π
interactions that results in the development of closely packed
molecular chains and uncovered substrate regions, (2)
electrostatic interactions between the large positive charge of
the S atom and the negatively Au substrate atoms which are
revealed by the Bader analysis (Table 1).
While we do not know if the (22 × √3) reconstruction is
lifted, the STM images of phase B closely resemble the
bipyridine patterns adsorbed on Au(111) that has been
assigned to physisorbed molecules on the reconstructed
surface.25 However, the DFT calculations result in similar
stability for 4MPySH on the (1 × 1) and (22 × √3) surfaces,
i.e., it is not possible to assess the state of the molecule−Au
interface.
When the potential reaches E > EPZC (the Au surface
becomes positive), the S−H bond of the 4MPyH species
should be broken, and the thyil radical is adsorbed on the Au
substrate. In this case, the Bader analysis indicates charge
transfer that results in a slightly negative charge on the S atom
and slightly positive charge on the Au atoms (Table 1). Note
that, in this case, the thione tautomer should be expelled from
the positively charged Au surface, but it can be immediately
readsorbed under the thiol form, yielding adsorbed thiyl radical.
On the other hand, the dense (5 × √3) chemisorbed lattice
(Figure 7b) exhibits the largest stability (the more negative γ
value); therefore, in principle, this phase appears as the best
candidate to organize the 4MPy molecules on the Au(111)
surface. However, there is a problem with the dense phases: the
surface coverage of the self-assembled 4MPy molecules before
the in situ STM experiments under electrochemical control
always yields smaller values than 0.15. Therefore, we need to
explain how the system rearranges the 4MPy molecules in
Figure 7. Optimized surface models for 4-mercaptopyridine on
Au(111). Right, top view and left side view. (a) (5 × √3) 4MPy*θ =
0.2. (b) (5 × √3) 4MPy*θ = 0.4. (c) (5 × √3) 4MPyNH θ = 0.4. (d)
4MPySH θ = 0.4, e) 4MPySH (22 × √3) θ = 0.41. (f) Side view of (22
× √3) surface model where only S atoms of 4MPySH are shown, the
surface corrugation is 0.3 Å. Key: yellow, Au atoms; green, S atoms;
gray, C atoms; blue, N atoms; white, H atoms. The unit cells of the
surface structures (white) are indicated.
F
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
order to have dense and diluted regions distributed on the
Au(111) surface in order to have an average surface coverage of
0.15. It is evident that the dense chemisorbed phase, which has
not only molecule−molecule but also strong molecule−
substrate interactions (large Eb), should be largely impeded
to reorganize the molecules. On the other hand, the weakly
adsorbed dense phases, which have very weak molecule−
substrate interactions (very small Eb), can easily rearrange the
initial material into dense patches optimizing molecule−
molecule interactions, thus resulting in uncovered regions as
shown in the STM images. However, this model also has a
weak point: the XPS data show that the initial 4MPy monolayer
is chemisorbed. A key piece of evidence about the nature of the
dense B phase is that the self-assembled monolayer remains
stable only if E < EPZC, where the gold surface is negatively
charged, i.e., it cannot be detected by XPS. Therefore, we
propose that the dense B phase corresponds to weakly
adsorbed 4MPy molecules (4MPySH, 4MPyNH) stabilized by
S(+)−Au(−) electrostatic interactions (Table 1). This
particular scenario can be obtained by first setting the
deposition time to 30 s in the self-assembly step and then
performing an electrochemical annealing in which the
molecules are first forced to be desorbed (close to the
hydrogen evolution region) and then readsorbed over the
surface by pushing the potential back again to 0.33 V. Note that
atomic H is effective to desorb alkanethiol molecules from the
Au(111) surface.33
When E > EPZC, the Au surface becomes positive and the
oxidative thiol adsorption takes place, resulting in S−H bond
breaking and charge transfer between the Au surface to the S
atom, leading to the chemisorbed C phase depicted in Figure
2c,d. This process implies a redistribution of molecules on the
Au(111)−(1 × 1) with larger inter row distances, perhaps
caused by a decrease in the repulsion of the surface dipoles
arising from the charge transfer. We suggest that the C phase,
which is incommensurate with the Au(111) surface, is a bridge
between the B and A phases since we can move from the C
phase to the A phase by simple reordering of the 4MPy
molecules. Note, however, that along with this complex process
it is also possible that chemisorbed and physisorbed molecules
coexist on the substrate surface.
The final question is why the Au samples immersed for 30s
and 2.5 min in the 4MPy containing solution, in spite of
showing the same surface chemistry and coverage, differ in their
behavior when the cathodic excursion is performed, in the first
case leading to the weakly adsorbed B phase and in the second
case remaining covered by the phase A. In order to approach
the subject, EC-STM images of freshly prepared substrates
(first scan-imaging) are taken into consideration.
After a few seconds of reaction, the interaction between the
substrate and the 4MPy produces a typical surface covered with
a homogeneous distribution of gold islands and no local
domains of self-assembly ordering (Figure 8A). If the selfassembly reorganization process is not forced to be stopped at
this point, the interaction between the molecules continues, the
gold islands begin to disappear, and, after a few minutes, some
domains of a stable (5 × √3) structure are formed (Figure
8B). Then we propose that at short immersion time the system
is unable to optimize the vdW interactions among the
molecules so that they can be easily desorbed during the
cathodic excursion relative to those present in the 2.5 min
immersed samples. This is not surprising, as it is well-known for
alkanethiolates SAMs that, while chemisorption is a fast
Figure 8. EC-STM images showing 4MPy/Au(111) self-assembled. 60
nm × 60 nm scan, Ebias = +400 mV, Itunn = 600 pA, HClO4 0.1 M. (A)
First scan of 30 s adsorption sample at EOCP = +0.330 V; (B) first scan
of 2.5 min adsorption sample at EOCP = +0.390 V.
process, optimization of molecule−molecule interactions is a
rather slow process.28
■
CONCLUSIONS
We have demonstrated the existence of a dense phase for 4MPy
SAMs prepared for short adsorption times and subject to a
cathodic excursion of potentials values close to the hydrogen
evolution reaction. This phase is stable at E < EPZC, where the
Au surface is negatively charged. We propose that this dense
phase is weakly adsorbed on the Au surface stabilized by
intermolecular π−π interactions and electrostatic forces
between the positively charged S atoms of the 4MPySH or
4MPyNH species with the negatively charged substrate surface.
The weakly adsorbed adlayer evolves into a closely related
chemisorbed lattice at E ≥ EPZC that can be easily transformed
into the well-known (5 × √3). By contrast, when the
adsorption time is increased to t > 2 min, only the (5 × √3)
lattice of chemisorbed 4MPy molecules is observed, irrespective
of the applied potential, i.e., the weakly adsorbed state is not
observed. The different behavior against a cathodic excursion
can be explained considering the optimization of the molecular
interactions at longer times that hinders the desorption−
reorganization needed to form the weakly adsorbed state since
both samples exhibit the same surface chemistry and coverage.
Our results allow a complete description and connections
between the different surface structures in terms of
thermodynamic stability.
G
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
■
(17) Klimeš, J.; Bowler, D. R.; Michaelides, A. J. Phys.: Condens.
Matter 2010, 22, 022201.
(18) Blöchl, P. E. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50,
17953.
(19) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188.
(20) Blöchl, P. E.; Margl, P.; Schwarz, K. In Chemical Application of
Density-Functional Theory; Brian, B., Laird, R. B. R., Ziegler, T., Eds.;
American Chemical Society: Washington, DC, 1996; Vol. 629, p 54.
(21) Kresse, G.; Joubert, D. Phys. Rev. B: Condens. Matter Mater. Phys.
1999, 59, 1758.
(22) Torres, D.; Carro, P.; Salvarezza, R. C.; Illas, F. Phys. Rev. Lett.
2006, 97, 226103.
(23) Reuter, K.; Scheffler, M. Phys. Rev. B: Condens. Matter Mater.
Phys. 2001, 65, 035406.
(24) Kolb, D. M.; Schneider, J. Electrochim. Acta 1986, 31, 929.
(25) Cunha, F.; Tao, N. J. Phys. Rev. Lett. 1995, 75, 2376.
(26) Sawaguchi, T.; Mizutani, F.; Taniguchi, I. Langmuir 1998, 14,
3565.
(27) Sawaguchi, T.; Mizutani, F.; Yoshimoto, S.; Taniguchi, I.
Electrochim. Acta 2000, 45, 2861.
(28) Vericat, C.; Vela, M. E.; Corthey, G.; Pensa, E.; Cortés, E.;
Fonticelli, M. H.; Ibañez, F.; Benitez, G. E.; Carro, P.; Salvarezza, R. C.
RSC Adv. 2014, 4, 27730.
(29) Boland, T.; Ratner, B. D. Langmuir 1994, 10, 3845.
(30) Ptasińska, S.; Stypczyńska, A.; Nixon, T.; Mason, N. J.;
Klyachko, D. V.; Sanche, L. J. Chem. Phys. 2008, 129, 065102.
(31) Lusteinberg, P. G.; Vericat, C.; Benitez, G. A.; Vela, M. E.;
Tognalli, N.; Fainstein, A.; Martiarena, M. L.; Salvarezza, R. C. J. Phys.
Chem. C 2008, 112, 11394.
(32) Wan, L. J.; Noda, H.; Hara, Y.; Osawa, M. J. Electroanal. Chem.
2000, 489, 68.
(33) Kautz, N. A.; Kandel, S. A. J. Am. Chem. Soc. 2008, 130, 6908.
AUTHOR INFORMATION
Corresponding Author
*E-mail: calvo@qi.fcen.uba.ar.
ORCID
Federico J. Williams: 0000-0002-6194-2734
Ernesto J. Calvo: 0000-0003-0397-2406
Present Address
́
Departamento de Quimica
de los Materiales, Facultad de
́
Quimica
y Biologia,́ Universidad de Santiago de Chile, Av.
Libertador Bernardo Ó Higgins 3363, Santiago, Chile
†
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
The authors acknowledge financial support from CONICET,
ANCPCyT and the University of Buenos Aires. F.T. thanks the
Teaching Program of Montevideo Group of Universities,
AUGM, for a visit grant to the University of Buenos Aires,
and Fondecyt 11130167 and "Proyectos Basales". P.C.
thankfully acknowledges MINECO (ENE2016-74899-C4-2-R,
AEI-FEDER-UE) and the computer resources provided by the
Computer Support Service for Research (SAII) at La Laguna
University.
■
ABBREVIATIONS
EC-STM electrochemical scanning tunneling microscopy;
4MPy 4-mercaptopyridine; SAMs self-assembled monolayers;
XPS X-ray photoelectron spectroscopy; DFT density functional
theory
■
REFERENCES
(1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides,
G. M. Chem. Rev. 2005, 105, 1103.
(2) Vericat, C.; Vela, M. E.; Benitez, G.; Carro, P.; Salvarezza, R. C.
Chem. Soc. Rev. 2010, 39, 1805.
(3) Häkkinen, H. Nat. Chem. 2012, 4, 443.
(4) Jin, Q.; Rodriguez, J. A.; Li, C. Z.; Darici, Y.; Tao, N. J. Surf. Sci.
1999, 425, 101.
(5) Poirier, G. E.; Pylant, E. D. Science 1996, 272, 1145.
(6) Pensa, E.; Rubert, A. A.; Benitez, G.; Carro, P.; Orive, A. G.;
Creus, A. H.; Salvarezza, R. C.; Vericat, C. J. Phys. Chem. C 2012, 116,
25765.
(7) Ramírez, E. A.; Cortés, E.; Rubert, A. A.; Carro, P.; Benítez, G.;
Vela, M. E.; Salvarezza, R. C. Langmuir 2012, 28, 6839.
(8) Cai, W. B.; Wan, L. J.; Noda, H.; Hibino, Y.; Ataka, K.; Osawa, M.
Langmuir 1998, 14, 6992.
(9) Wan, L. J.; Terashima, M.; Noda, H.; Osawa, M. J. Phys. Chem. B
2000, 104, 3563.
(10) Wan, L.-J.; Hara, Y.; Noda, H.; Osawa, M. J. Phys. Chem. B 1998,
102, 5943.
(11) Baunach, T.; Ivanova, V.; Scherson, D. A.; Kolb, D. M. Langmuir
2004, 20, 2797.
(12) Zhou, W.; Baunach, T.; Ivanova, V.; Kolb, D. M. Langmuir
2004, 20, 4590.
(13) Kǔcera, J.; Gross, A. Langmuir 2008, 24, 13985.
(14) Kresse, G.; Hafner, J. Phys. Rev. B: Condens. Matter Mater. Phys.
1993, 48, 13115.
(15) Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6, 15.
(16) Dion, M.; Rydberg, H.; Schröder, E.; Langreth, D. C.;
Lundqvist, B. I. Phys. Rev. Lett. 2004, 92, 246401.
H
View publication stats
DOI: 10.1021/acs.langmuir.7b01627
Langmuir XXXX, XXX, XXX−XXX